Comment 81 for bug 375148

Then you weren't here when I first really joined into this discussion after
purchasing 6 additional modems spanning multiple vendors and setting aside two
BILLABLE days to assist with the debugging so I could once again fax a few
contracts. That is the two days we got two famous releases, the first didn't
compile and the second didn't install no matter which modem I put in my
machine.

Please read an _ENTIRE_ message thread before stating others have done and are
doing nothing.

If you want, unsubscribe me from this bug. It is obvious that no Linux
distribution will support a functioning built in modem past either a single
installation or a single .n release of the kernel. The Intel modem software
works with -14 and no other version of the kernel, which I guess is better
than this, but rebooting to a much older kernel just to send a fax is going to
cause a severe problem with other applications at some point.

The architecture which the Linux community has chosen to use supporting modems
is quite simply broken. Until that fatal design flaw is fix, and until someone
spends a lot of thankless time reverse engineering the modem code, Linux will
never again support modems or faxing. I have been getting by paying $1.99 per
fax using an Internet faxing service.

After the end of April when the new LTS comes out, I will be switching to 64-
bit once again so this bug will be immaterial.

I only respond now because my 20+ years in IT tells me people are simply
spinning their wheels on this and the Linux community as a whole should simply
state they cannot compete with Windows when it comes to faxmodem support.
Either break the law and disassemble the object code provided and write your
own high level code from that or punt.

What is going to happen if Ahmed magically gets the stars to align and
produces a version which works perfectly for a single 32-bit user? A shiny
new -NN kernel header version will come out and it will cease to work. I've
watched this for over two years with the Intel stuff. It's no different here.
When Ahmed says he got it working, he got it working for the -NN kernel header
version he was using, which apparently is different that what everybody else
following this thread is using. Today -20 came out, or at least I installed
it today because Ubuntu pushed it out. Is that the version he tested with?
Probably not.

The point in time where the Linux engineers/community choose to load faxmodem
drivers is simply flawed. I've had 9 kernel updates since I installed Karmic.
All of my USB devices still work...sans the hot key for the KVM which hasn't
worked with any version of this release once the GUI is loaded. All of my
disk drives work with all versions of the kernel. The Intel modem I installed
after the announcement came out only works with -14. The other modems I
bought for helping with this problem don't work with any kernel version on
this release of Ubuntu, but the one worked with both OpenSuSE and the previous
32-bit version of Ubuntu.

My only real question right now is what USB faxmodem does this other guy have
and has he gotten it working sending faxes with 64-bit Ubuntu?

On Monday 22 March 2010 05:58:21 pm Rolf Leggewie wrote:
> Seasoned_geek, you mention a few of the frustrations that I remember
> myself. But to be fair to Ahmed's work, installation or compilation
> problems are something I haven't seen for quite a while now. Please
> acknowledge the progress being made instead of perpetuating statements
> that are no longer true for quite a while now. You continue to make
> frequent and far-reaching demands on other people, but I've yet to see a
> single contribution from you towards solving this problem. Not quite
> what I would expect from a "seasoned geek".
>
> If Ahmed did indeed get the device to be created correctly, that's
> already a good step forward. If that modem then dialed out and
> connected, that's even better news. Giving Ahmed's history of claiming
> things are fixed when in fact they weren't, I guess it's normal that
> people have now grown suspicious and I encourage Ahmed to explain in
> more detail what he did test and how his successes were instead of
> merely closing a ticket with "package XY in Debian fixes it" (and then
> doesn't deliver on that promise once more).
>
> Let's see if we can understand why Ahmed can apparently do some things
> that I still can't. Him running Karmic with an older kernel may be one
> of the reasons.
>

--
Roland Hughes, President
Logikal Solutions
(630)-205-1593 (cell)
http://www.theminimumyouneedtoknow.com
http://www.infiniteexposure.net
http://www.logikalsolutions.com